RADIOGRAPHIC ABNORMALITIES IN TUBERCULOSIS AND RISK OF COEXISTING HUMAN-IMMUNODEFICIENCY-VIRUS INFECTION - RESULTS FROM DAR-ES-SALAAM, TANZANIA, AND SCORING SYSTEM

Citation
N. Mlikacabanne et al., RADIOGRAPHIC ABNORMALITIES IN TUBERCULOSIS AND RISK OF COEXISTING HUMAN-IMMUNODEFICIENCY-VIRUS INFECTION - RESULTS FROM DAR-ES-SALAAM, TANZANIA, AND SCORING SYSTEM, American journal of respiratory and critical care medicine, 152(2), 1995, pp. 786-793
Citations number
14
Categorie Soggetti
Emergency Medicine & Critical Care","Respiratory System
ISSN journal
1073449X
Volume
152
Issue
2
Year of publication
1995
Pages
786 - 793
Database
ISI
SICI code
1073-449X(1995)152:2<786:RAITAR>2.0.ZU;2-6
Abstract
First, we evaluated the age profile and chest radiographic abnormaliti es in 146 patients from Dar-es-Salaam, Tanzania, with new-onset intrat horacic tuberculosis (pulmonary, pleural, or hilar/mediastinal adenopa thy), to identify features that were associated with human immunodefic iency virus (HIV) seropositivity or seronegativity; then, we combined these data with those from a companion investigation in Burundi to dev elop a simple scoring system to predict HIV serologic status. Using ag reed-upon criteria and simplified reporting forms, initial chest radio graphs were reviewed by three readers, first independently and then at a consensus conference. Of the 146 patients, 80 (55%) were HIV seropo sitive and 66 were seronegative. More seropositive than seronegative s ubjects were 31 to 40 yr old (p = 0.03). Because the radiographic char acteristic: of the two serologic groups were similar in Tanzania and B urundi, we combined the data for stepwise logistic regression that rev ealed four highly significant variables: age, small lesions, location, and lymphadenopathy. From these, we obtained an equation to calculate the probability that a given tuberculosis patient was HIV seropositiv e and then we derived a scoring system that in its simplest form (thre shold) predicted serologic status correctly in 68.1% of patients; a gr aded scale was even more accurate in the high (89.1%) and low (82.6%) ranges. This scoring system should be useful when serologic testing is unavailable or refused.